Research Paper link: https://drive.google.com/file/d/19tlUoon3kCyNZSuh6IMt25eb14S5LERy/view?usp=sharing
The Investment Manager is designed to help users achieve the best possible returns on their investments by tailoring portfolios according to their risk preferences. By leveraging advanced machine learning techniques and financial theories, the application aims to offer optimized investment strategies.
Table of Contents 🗓
Check requirements.txt
The forementioned libraries should be installed in order for the code to run properly. All the libraries can be downloaded to the specified version using the following command:
- npm
npm install npm@latest -g
- library
npm install library_name_here -g
After downloading the project, install the required libraries. In the console run the command: "streamlit run nus.py"
- Indentify possible datasets
- Balance the data
- Create meta.csv (labels)
- Create a preprocessing pipeleine
- Data cleaning and preparation
- feature extraction and selection
- Implementation of RNN
- Develop and train LSTM
- Optimise the performance
- Apply Markowitz Portfolio Theory
- Risk-return analysis
- Portfolio optimisation
- Backtesting
Rishit Saraf - @rishitsaraf - [email protected]
Research Paper link: https://drive.google.com/file/d/19tlUoon3kCyNZSuh6IMt25eb14S5LERy/view?usp=sharing
Deep learning with long short-term memory networks for financial market predictions by Thomas Fischer and Christopher Krauss
Long Short-Term Memory Neural Network for Financial Time Series by Carmina Fjellström
Using Market News Sentiment Analysis for Stock Market Prediction by Marian Pompiliu Cristescu et al.
Robo-advisor using genetic algorithm and BERT sentiments from tweets for hybrid portfolio optimisation by Edmund Kwong Wei Leow et al.
The Black-Litterman Approach: Original Model and Extensions by Attilio Meucci